With the development of the Internet, information distribution through web pages has become an important social key technology. Therefore, it is desired that web pages be easy to view even for users (viewers of web pages) having trouble with their eyesight, such as visually impaired people and senior citizens. Measures have been taken for readability of a web page, such as providing a style sheet for each user on a client or providing a transcoding system on a network. However, it will be more advantageous as fundamental measures that an original web page be designed to be easy to view.
Accessibility checkers have been used as a technique by which web page creators can improve accessibility of web pages. An accessibility checker analyzes an HTML (Hypertext Markup Language) document and checks its accessibility in accordance with a guideline established by WAI (Web Accessibility Initiative) of W3C (World Wide Web Consortium). Modifying the web page based on the result of this checking can produce a web page that conforms to the WAI guideline.
As to techniques not limited to web pages but generally related to image display using a computer, there are other techniques in the art regarding the eyesight of visually impaired people and senior citizens. One technique involves generating an image and subjecting it to image processing corresponding to a visually impaired person, an aged person, or any of various symptoms of visual impairment (such as weakening eyesight, astigmatism, diplopia, contrast, scattering, tunnel vision, visual field defect, etc.) for displaying it on a screen. Another technique discloses means for not only simulating and displaying how a certain image is viewed by senior citizens, but also making modifications by pixel-by-pixel image processing so that the image becomes easy to view for them.
As described above, it would be desirable to create an easily viewable image for displaying on a computer, for example a web page. However, the technique of using the accessibility checker to check the accessibility of a web page has the following problems. One first problem is that the analysis of an HTML document has a limitation. The checker cannot provide a specific direction about how to modify the image, so that the analysis predominantly relies on human eyes in the end. For example, the above mentioned Bobby can exactly indicate a position having a problem that “there is an image without an ALT attribute”, because the problem can be found simply by checking tags. However, it cannot indicate a position having a problem that “the contrast is insufficient between the foreground color and the background color.” Therefore, checking problems such as the latter has to rely on a web page creator's eyes. Such check items left to human eyes actually occur in large numbers, which makes efficient checking quite difficult.
Another problem is that, even if a web page is modified for all its check items to ensure that the web page conforms to the WAI guideline, the web page may not actually become easily viewable by a user having problem with his/her eyesight.
Because the guideline is intended to address various disabilities like the highest common factor and is limited to a feasible number of items, it naturally fails to accommodate every web page to perfection.
The fundamental cause of these problems with the accessibility checker lies in that the readability to be achieved when an image is displayed on a screen is checked on an HTML document (text) basis.
On the other hand, prior art apparently simulates how a generated image is viewed by a user having problem with his/her eyesight and displays the result on a screen. However, these techniques only performs simulation for purposes such as virtual experience education and do not take into consideration measures to modify the image so that the user having problem with his/her eyesight can easily view the image.
Another technique has means for modifying the image to be easily viewed by senior citizens, but the technique cannot be applied to web page creation because the modification is based on pixel-by-pixel image processing. The reason is that, for a web page, a web page creator can modify only an HTML document (text) but not a raster image resulted from rendering the HTML document.
Also, the prior art apparently is designed to modify an original image uniformly based on the age of an expected aged user and to display the modified image to the user. It does not indicate which and how portions of the original image are hard to view. In other words, it does not provide an advice or hint about how to configure the original image.
Therefore, if the original image has a bad quality, the readability of the image cannot be improved beyond the limitation of the modification means. For example, if characters in the image are too small to be legible, pixel-by-pixel image processing (such as edge emphasis or contrast improvement) can hardly improve the legibility.